Advanced Skills in Modern Radiotherapy 2018

Advanced Skills in Modern Radiotherapy

06-10 May 2018 - Rome, Italy

Speakers

Course Director • Rianne de Jong

Faculty • Elizabeth Forde • Mirjana Josipovic (not present) • Martijn Kamphuis • Jose Lopez • Peter Remeijer • Sofia Rivera

Guest Lecturers • Maaike Milder • Marco Schwarz • Local lecturers: Veronica Pollutri & Francesco Cellini

Programme: DAY 1

Time

Description

Speaker

09.00 – 09.15 Welcome & Introduction of teachers

R.de Jong

09.15 – 09.45 RTT’s Perspective on modern radiation therapy

R. de Jong

09.45 – 10.15 Patient preparation and positioning

M. Kamphuis

10.15 – 10.45 Coffee break

10.45 – 11.30 Pre-treatment Imaging Modalities

P. Remeijer

11.30 – 12.15 Delineation Target Volumes

S. Rivera

12.15 – 13.00 Delineation Organs at Risk

E. Forde

13.00 – 14.00 Lunch break

14.00 – 14.15 Workshop on delineation of OAR: Introduction to the software

S.Rivera / E. Forde / P. Remeijer

14.15 – 15.30 Workshop on delineation of OAR

S.Rivera / E. Forde / P. Remeijer

15.30 – 16.00 Coffee break

16.00 – 17.00 Workshop on delineation of OAR

S.Rivera / E. Forde / P. Remeijer

Programme: DAY 2

Time

Description

Speaker

08.30 – 09.00 Errors and Margins

P. Remeijer

09.00 – 09.30 In room imaging modalities

M. Kamphuis

09.30 – 10.00 Correction Strategies

P. Remeijer

10.00 – 10.30 Coffee break

10.30 – 12.15 Workshop on margin calculation: part I

P. Remeijer

12.15 – 13.15 Lunch break

13.15 – 13.45 Motion Management

P. Remeijer

13.45 – 14.15 Image registration

P. Remeijer

14.15 – 14.45 Treatment Planning I

E. Forde

14.45 – 15.15 Coffee break

15.15 – 15.45 Treatment Planning II

E. Forde

15.45 – 16.15 Clinical rationale for IGRT

J. Lopez

16.15 – 16.45 Workshop on margin calculation: part II

P. Remeijer

Programme: DAY 3

Time

Description

Speaker

08.30 – 10.15 Lower Abdomen: Prostate & cervix (6x 15 min)

Faculty

10.15 – 10.45 Coffee break

10.45 – 12.30 Thorax: Lung and breast (6x 15min)

Faculty

12.30 – 13.30 Lunch break

13.30 – 14.15 Image registration and Evaluation: Part I (CBCT XVI)

R. de Jong

14.15 – 15.00 Image registration and Evaluation: Part II (CBCT Varian)

E. Forde

15.00 – 15.30 Coffee break

Break up sessions Image registration and evaluation Varian & Elekta

15.30 – 17.15

Programme: DAY 4

Time

Description

Speaker

09.00 – 09.30 Recap Registration Workshop

R. de Jong

09.30 – 11.15 Head&Neck (3x 15min) / Brain (3x 15min)

Faculty

11.15 – 11.45 Coffee break

11.45 – 12.15 Implementing and managing IGRT

M. Kamphuis

12.15 – 13.00 Who is doing what in radiation therapy - interactive -

R. de Jong

13.00 – 14.00 Lunch break

14.00 – 15.30 Workshop: Safety issues and prospective risk analysis

M. Kamphuis

15.30 – 16.00 Coffee break

16.00 – 16.30 Cyberknife – Skype lecture

M. Milder

16.30 – 17.00 Error management

P. Remeijer

Programme: DAY 5

Time

Description

Speaker

08.30 – 10.00 Theory & Workshop: Plan of the day

R. de Jong

10.00 – 10.30 Incident management

M. Kamphuis

10.30 – 11.00 Coffee break

11.00 – 11.30 Adverse Event Reporting and the Role of the RTT

E. Forde

11.30 – 12.00 Protons

M. Schwarz

12.00 – 12.30 MR guided treatment

Local lecturer

12.30 – 13.30 Wrap-up & Closure

Faculty

Patient Preparation and Positioning

Martijn Kamphuis MSc MBA

(Slides: Rianne de Jong) Academic Medical Center, Amsterdam Prague 2017

m.kamphuis@amc.nl

Aim of Patient preparation and positioning

Minimize the difference in patient position 1. between simulation and treatment sessions 2. during the treatment session Maximize the distance between target volume and organs at risk

Tools: •

Immobilization and fixation

Patient compliance

•

3

Tools of Patient preparation and positioning

Immobilization Daily set-up reproducibility and stability through the use of fixation or aiding devices

4

Expectation management

• This aim of this talk is not to show the best devices

• Understanding the rationale behind it

• Choice for device will be based on: ➢ Economics ➢ Local availability ➢ Literature ➢ Experience

• Link to important review at the end of the .ppt

Tools of Patient preparation and positioning

Patient compliance

– Information and education • Using photo books, DVD’s, folders etc. • Tour through department – Psychological support to minimize fears – Practical session in case of SBRT – Medication • Pain control

6

Minimize the difference in patient position

Minimize the difference in patient position 1. between simulation and treatment sessions 2. during the treatment session Maximize the distance between target volume and organs at risk

Tools: •

Patient compliance

Immobilization and fixation

•

7

Aim of Patient preparation and positioning

Minimize the difference in patient position between simulation and treatment sessions: inter -fraction motion

Tools: Patient compliance: •

Pelvic patients using diet / drinking protocol

Immobilization and fixation: •

Head&Neck using head support

Lung using 4D CBCT.

•

8

Prostate patients

Reconstructed

CBCT

11

Prostate patients

To improve image quality: Dietician

– Mild regimen of laxatives – Diet

Fixed treatment times

12

Prostate patients

gas

faeces moving gas

no diet

68% 61%

45%

with diet

42% 23%

22%

• reduced percentage of faeces and gas

• reduced percentage of moving gas, hence improved image quality

M. Smitsmans

13

Prostate patients

Lips et al. Ijrobp 2011 • 739 patients without diet, 205 patients with diet • Diet instructions on leaflet • No reduction of intrafraction movement

McNair et al. 2011 • 22 patients using questionaires

• Rectal filling consistency not improved • Diet + fixed treatment times, no laxatives

Conclusion: • Drinking and dietery protocol are needed for clear patient communication BUT • Won’t solve the whole problem of intra/interfraction motion (additional tools are needed)

14

Aim of Patient preparation and positioning

Minimize the difference in patient position between simulation and treatment sessions: inter -fraction motion

Tools: Patient compliance: •

Pelvic patients using diet / drinking protocol

Immobilization and fixation: •

Head&Neck using head support

Unfortunate differences

•

15

Head&Neck patients: head support

Rigid registration BSpline registration Deformation field

Coronal

Sagittal

16

Head&Neck patients: head support

• Reduction of the average difference between fractions in set up of the bony anatomy. • Reduction in the difference of the shape of the bony anatomy between fraction.

A. Houweling

Creating unfortunate differences

• Between CT and treatment

Example 1: Look for differences..

Example 2: Respiratory monitoring system

• 4D CBCT scans with and without oxygen mask • 3D tumor motion was assessed for tumor mean position and amplitude

J. Wolthaus, M. Rossi

20

Respiratory monitoring system

With oxygen mask

Without oxygen mask

AP (cm) CC (cm) LR (cm)

LR (cm)

CC (cm)

AP (cm)

0.18

0.23

0.23

0.15

0.21

0.22





0.06 0.16

0.03 0.19

0.00 0.19



0.18 0.04

0.17 0.08

0.20 -0.09



Mean

Mean

No significant difference in tumour mean position

J. Wolthaus, M. Rossi

21

Respiratory monitoring system

1.8

Oxygen Mask No Mask

1.6

1.4

1.2

1

0.8

0.6

Breathing Amplitude [cm]

0.4

0.2

0

1

2

3

4

5

6

7

8

9

Patient

M = 29%, SD = 19%, p = 0.0017

Difference in breathing amplitude!

J. Wolthaus, M. Rossi

22

Deformable registration decreases the need for good immobilization

A.True B.False

Aim of Patient preparation and positioning

Minimize the difference in patient positioning during the treatment session: intra -fraction motion

Tools: Increasing patient compliance: • Immobilization and fixation: •

Practical session SBRT

Lung using 4D CBCT.

26

Practical session

In case of hypofractioned RT: • Patient visit the linac • Session is completely performed but no Gray’s are given

Advantages: • Patient gets acquinted with workflow • Set-up accuracy can be assesed: ➢ is the intra# motion acceptable? • Is it do able for the patient? • Is the image quality sufficient? • Precautions can be made: ➢ Pain/stress relief ➢ Additional margins/replanning

Stability with prolonged treatment time

Hypo fractionated lung

On-line lung tumor match with CBCT: 3 x 18 Gy (first protocol design without arc therapy and inline scanning)

Aligning the patient:

5 min 4 min 5 min 3 min 4 min 1 min

First CBCT scan:

Registration:

Manual table shift: Second CBCT scan: Evaluation CBCT scan:

Beam delivery:

25 min

Post treatment CBCT scan:

4 min

28

Stability with prolonged treatment time

Antoni van Leeuwenhoek Hospital

29

Stability with prolonged treatment time

Antoni van Leeuwenhoek Hospital

30

Stability with prolonged treatment time

59 Patients, 3 fractions per patient

LR (mm)

CC (mm)

AP (mm)

GM

0.2

0.6

-0.6

Residual Inter- fraction

0.8

0.8

1.0



1.1

1.1

1.4



GM

0.0

1.0

-0.9

Intra-fraction

1.2

1.3

1.9



1.2

1.4

1.7



Antoni van Leeuwenhoek Hospital

31

Intrafraction motion is the motion of a patient within a session

A. True B. False

Patient compliance won’t impact intrafraction motion

A. True B. False

Minimize the difference in patient position

Minimize the difference in patient position 1. between simulation and treatment sessions 2. during the treatment session Maximize the distance between target volume and organs at risk

Tools: •

Immobilization and fixation

Patient compliance

•

34

Minimize the difference in patient position

Maximize the distance between target volume and organs at risk

Tools: Immobilization and fixation:

• Bellyboard for pelvic patients

Patient compliance:

• Breath hold for breast patients

35

Belly board pelvic patients

Belly board

36

Belly board pelvic patients

Rectum patients

Das et al, 1997

37

Breath hold for breast patients

Normal inspiration

Deep inspiration

J. Sonke

38

Essential: education & compliance

Patient preparation and immobilization aims at:

A. Minimizing patient compliance B. Maximizing intrafraction motion C. Minimizing inter- and intrafraction motion D. Decreasing the distance between PTV and OAR’s

Conclusion

The first step in radiation therapy is to minimize

• the difference in patients anatomy and set-up between CT en treatment • the difference in patients anatomy and set-up between treatment days

and to maximize

• patient stability • the distance between target volume and organs at risk

41

Conclusion

The first step in radiation therapy is to minimize

• the difference in patients anatomy and set-up between CT en treatment • the difference in patients anatomy and set-up between treatment days

and to maximize

• patient stability • the distance between target volume and organs at risk

42

Conclusion

https://espace.cern.ch/ULICE-results/Shared%20Documents/D.JRA_5.1_public.pdf

‘Recommendations for organ depending optimized fixation systems’

Pre-treatment imaging

Mirjana Josipovic Dept. of Oncology, Rigshospitalet & Niels Bohr Institute, University of Copenhagen Denmark

Advanced skills in modern radiotherapy May 2018

Intended learning outcomes

• Illustrate the importance of a particular pre-treatment imaging modality for radiotherapy • Comprehend the additional value of applying combined information from several imaging modalities for radiotherapy planning

• Identify uncertainties of pre-treatment imaging modalities

Pre-treatment imaging for radiotherapy

• CT: computed tomography

• PET: positron emission tomography

• MR: magnetic resonance

Do you have experience with…?

A. CT B. PET/CT

C. PET D. MR E. PET/MR F. None of the above

Multiple answers possible!

Which imaging modalities do we need for modern state of the art radiotherapy?

A. CT B. PET C. MR

D. CT & PET E. CT & MR F. PET & MR G. CT & PET & MR

CT chronology

• 1917 mathematical grounds for CT reconstruction

• 1971 first clinical CT

• 1990 spiral CT • 1993 dual slice • 2003 32-slice

• Today : ultrafast volume-scanning dual source, dual energy

80x80 matrix 5 min rotation time

1024x1024 matrix < 0.3 s rotation time

PET chronology

• 1930’s radioactive tracers

• 1953/66 multidetector device

Wagner et al. 1998

• 1975 back projection method for PET

• 1979 fluorine 18 deoxy glucose (FDG)

• 2000 PET/CT “medical invention of the year”

MR chronology

• 1937 nuclear magnetic resonance

• 1956 Tesla unit • 1972 Damadian invention

• 1977 first MR scan

• 1993 functional MR

CT

MR

PET

T1

T2

flair

CT

MR

PET

What do we see?

• Morphology

(patologic) anatomy

➢

CT, MR

Tumour metabolism Perfusion Organ function

• Biological processes

➢

PET, MR

Diagnostic imaging vs RT imaging

• Diagnostic

➢

What is this?

• RT planning

➢

Where is this?

Why we need CT

CT numbers = Hounsfield units

The grey tones on the CT image represent the attenuation in every pixel/voxel

The grey tones are expressed in Hounsfield units (HU) – CT numbers:

μ

– μ

obj water HU = –––––––– x 1000 μ water

Luft ~ -1000 HU Vand ~0 HU Knogler >1000 HU

Hounsfield units → electron density

Necessary for dose calculation

Calibration curve needed for each applied kV

How well can we trust the imaging information?

Image artifacts

Definition : Systematic deviation between the HU in the reconstructed image and the objects correct attenuation’s coefficient

• Partial volume artefacts • Streak artefacts • Ring artefacts • Motion artefacts • Noise

Partial Volume artefacts

Oxnard et al. JCO 2011

Variability of Lung Tumor Measurements on Repeat Computed Tomography Scans Taken Within 15 Minutes

For a lesion measuring 4 cm, CT variability can lead to measurements from 3.5 to 4.5 cm

Streak artefacts

Metal artifact reducton sw

• Dual Energy CT (DECT)

▪

Used two different X-ray energies

▪

“Virtual monochromatic” scans

• Iterative metal artifact reduction software

▪

MAR, iMAR, O-MAR...

MAR - impact on dose planning

Dose calculation for 10 patients with iMAR – No difference in dose compared to manual override

Images courtesy of Laura Rechner, Rigshospitalet

MAR- impact on contouring

• Head and neck contouring by a radiation oncologist

Images courtesy of Jeppe Friborg, Rigshospitalet

MAR combined with dual energy scan

• Which images do radiologists & oncologists prefer?

120 kVp 70 keV 130 keV

5

6

4

No MAR

120 kVp iMAR 70 keV iMAR 130 keV iMAR

1

3

2

MAR

Kovacs et al. RO 2018

MAR combined with dual energy scan

• Which images do radiologists & oncologists prefer?

No MAR

120 kVp 70 keV 130 keV

4

5

6

120 kVp iMAR 70 keV iMAR 130 keV iMAR

1

2

3

MAR

Kovacs et al. RO 2018

Imaging for RT planning

• Has to be precise • Has to provide safe judgment of the extent of the disease

• CT images are base for treatment planning

BUT • On CT, it can be difficult to discriminate vital tumour tissue from scar tissue, oedema, atelectasis, surrounding soft tissu…

• CT can not stage correctly ➢ detect small metastases ➢ detect distant metastases

Added value of PET CT for radiotherapy

• Improved delineation consistency • Improved staging

Which sites do you plan with PET/CT?

A. Head/neck B. Lung C. Lymphoma D. Esophagus

E. Gyne F. Other G. We don’t use PET/CT

Multiple answers possible!

Improved delineation consistency

CT based

PET/ CT based

Steenbakkers IJROBP 2006

Impact of PET in lung cancer RT

• Change in target definition: in 2 out of 5 patients

• Change in treatment intent: in 1 out of 5 patients

PET imaging of brain tumours

FDG-PET

MR

FET-PET

• 18F-Fluoro-Ethyl-Tyrosin (FET), aminoacid uptake

BD Kläsner et al. Expert Rev. Anticancer Ther 2010

PET imaging of hypoxia with FMISO

• Hypoxia area is associated with high risk of locoregional failure

Thorwarth BJR 2015

Pitfalls

• FDG is not specific ➢

Not all ”hot-spots” are malignant

• Motion blurs the FDG uptake ➢

Courtesy of TL Klausen

Is it a small lesion, with high degree of motion and high SUV uptake? ➢ Is it a large lesion, without motion and low SUV uptake?

Courtesy of M Aznar

Free breathing

Breath hold

Added value of MR imaging for RT

• Superior soft tissue contrast

Which sites do you plan with MR?

A. Brain B. Head/neck C. Gyne D. Prostate

E. Liver F. Spine G. Other H. We don’t use MR

Multiple answers possible!

Prostate cancer

MR

CT

Functional imaging with MR

CT

T2

DCE (ktrans)

ADC

DCE = dynamic contrast enhanced • high signal due to increase in capilar permeability

ADC = apparent diffusion coefficient • lack of signal due to high cell density

Pitfalls

• Geometric distortion

Schmidt & Payne PMB 2015

• No direct relation with electron density

➢

CT atlas corregistration

➢

MR segmentation

MRI artifacts can cause invisible geometrical errors!

Water fat shift: Made visible by introducing a small read out gradient, but reversed in both images

Difference image

→ Relative position of bone and tumor geometrically incorrect

Courtesy U. van der Heide

Registration

• Planning and image guidance is CT and CBCT based • Delineation often based on MRI or PET

→ Registration error = Delineation error!

• Be careful with registrations – especially deformable

Anything can be deformed in anything else… But is it true?

Challenge of multi modality imaging

Daisne et al. Radiology 2004

Which imaging modalities do we need for modern state of the art radiotherapy?

A. CT B. PET C. MR

D. CT & PET E. CT & MR F. PET & MR G. CT & PET & MR

Conclusion (1)

• Illustrate the importance of a particular pre-treatment imaging modality for radiotherapy

➢ CT is needed for calculation of dose distribution ➢ PET adds value for staging, distinguishing tracer avid areas/volumes

➢

MR increased soft tissue contrast

Conclusion (2)

• Comprehend the additional value of applying combined information from several imaging modalities for radiotherapy planning

➢

More reproducible target definition

➢

More precise target definition

➢

Optimal treatment strategy

Conclusion (3)

• Identify uncertainties of pre-treatment imaging modalities

➢

Artefacts in images

➢ Differences in (spatial) info on each modality

Please score this lecture

A. Poor B. Sufficient C. Average D. Good E. Excellent

comments can be written on the paper form

TARGET VOLUME DELINEATION

Sofia Rivera, MD, PhD Radiation Oncology Department

Gustave Roussy Villejuif, France

Advanced skills in modern radiotherapy May 06, 2018

What is the weakest point in our modern radiotherapy treatment chain?

A. Dose calculation? B. Positioning uncertainties? C. Contouring uncertainties? D. Quality control of the treatment machine? E. Patient changes (weight loss, movements…)? F. RTTs?

G. Physicists? H. Physicians?

Learning outcomes

• Understand why heterogeneity in contouring is a major weak point in modern radiotherapy

• Discuss the challenges in contouring target volumes

• Identify skills required to delineate target volumes

• Identify tools for improving learning in delineation

• Identify adequate imaging modalities according to the target to delineate

• Discuss the impact and consequences of inaccurate delineation of target volumes

Delineation: one of the links in the treatment chain

Why is delineation important?

• Radiotherapy planning is nowadays mostly based on delineation

• Constraints for dose distribution are used

• DVH are calculated based on the contours

• Field arrangements are becoming more complex

• An error in contouring will therefore translate in a systematic error all along the treatment and may have consequences: ➢ Jeopardizing treatment efficacy ➢ Impacting treatment toxicity

Do we need to improve?

How can we answer that need ?

➢ Adequate imaging, training and use of contouring guidelines are the main strategies to minimize delineation uncertainties ( Petrič et al 2013)

➢ Establishing and using consensus and guidelines have shown to reduce heterogeneity in contouring

NIELSEN et al 2013

Participants in the FALCON-IAEA study

14 centers from 13 countries that recently shifted from 2D to 3D

Structure of the FALCON-IAEA study

HNSC C

Lung

Cervix

Participants characteristics

Characteristic

Frequency

Female

39/57 (68%)

• 60 physicians were invited

Public hospitals

45/57 (80%)

Qualified specialist

44/57 (77%)

• 57 joined and delineated

Rutinely use 3D confomal RT

50/57 (88%)

Use IV contrast

34/57 (60%)

Image fusion

35/57 (61%)

Use intl. guidelines/atlas

52/57 (91%)

Regular peer-review 26/57 (46%) Confident radiology 39/57 (68%) Confident contouring 51/57 (89%)

Increased homogeneitey to reference contour – also 6 months after teaching

Level II-IV, Neck CTV-T, Cervical cancer

Before teaching

During teaching 6 mo after teaching

Before teaching

During teaching 6 mo after teaching

Did you know before this course that ESTRO provides a platform for hands on exercises on contouring?

A. YES B. NO

Inter-observer variability in contouring Examples of participant contours from ESTRO FALCON workshops. A: CTV breast, B: GTV Brain tumour, C: CTV prostate and D: GTV cervix cancer

B

A

C

D

Does heterogeneity in RT matters?

• Bioreductive agent

• Radiosensitizer in hypoxia

RT + CDDP

Multicentric international Randomized phase III 853 locally advanced H&N patients

RT + CDDP + Tyrapazamine

Hypoxia radioresistance

No benefit in overall survival

Rischin D et al. JCO 2010;28:2989-2995

©2010 by American Society of Clinical Oncology

But… Trial quality control

Peters L J et al. JCO 2010;28:2996-3001

©2010 by American Society of Clinical Oncology

Impact of radiotherapy quality

Peters L J et al. JCO 2010;28:2996-3001

©2010 by American Society of Clinical Oncology

How to improve?

• Need for a common language: ICRU

• Need for delineation guidelines and anatomical knowledge

• No absolute truth so need to specify according to which guidelines we contour

• Heterogeneity in understanding/interpreting the guidelines

• Need for teaching in contouring

• Need for evaluation in contouring

ICRU Guidelines (ICRU50): volume definition • Volumes defined prior/ during treatment planning: ➢ Gross Tumor Volume (GTV) ➢ Clinical Target Volume (CTV) ➢ Planning Target Volume (PTV)

➢

Organs At Risk (OAR)

➢

Treated Volume

➢

Irradiated Volume

• Volumes might be redefined during treatment for adaptive RT

Tumor Gross Volume: GTV

• Macroscopic tumor volume visible or palpable

• Includes: ➢

Primary tumor

➢

Macroscopically involved lymph nodes

➢

Metastases

What is your GTV when the tumor has been removed surgically like in a lumpectomy for breast cancer?

A. Whole breast B. Tumor bead C. Surgical clips D. There is no GTV

Tumor Gross Volume: GTV

• GTV is defined based on clinical data (inspection, palpation) and imaging (CT, MR, US, PET depending on it’s relevance for the tumor site)

• Definition of the GTV allows for TNM classification of the disease

• Definition of the GTV allows for tumor response assessment

• Adequate dose to GTV is therefore crucial for tumor control

Tumor Gross Volume: GTV

24

Which contour is the GTV?

A/ Blue B/ Red C/ Green

Which contour is the GTV?

A. Blue B. Red C. Green

Which one is the GTV?

Are you sure about your GTV????

PET scans in delineation of lung cancer

• FDG-PET has an established role in contouring NSCLC

• Changes the tumor GTV in about 30–60% of patients

• Changes the nodal GTV in 9–39% of patients mainly through detection of occult metastases not seen on CT, lowering the risk of nodal recurrences

Tumor Gross Volume: GTV

• Adequate high quality imaging is a key point

Images from the FALCON platform; case Lung PET: Vienna 2013

Clinical Target Volume: CTV

• Includes GTV + microscopic extension of the tumor

• Volume to adequately cover to ensure treatment efficacy weather treatment is delivered with a curative or a palliative intent

• CTV delineation is based on local and loco regional capacity/probability of extension of the tumor

• Includes potential micromets surrounding the GTV

• Includes potential micromets in tumor’s drainage territory

CTV

Clinical Target Volume: CTV

• High quality images are a key point for CTV delineation as well • Margins adapted to anatomical boundaries

GTV and CTV

• Definition based on:

➢

Anatomy

➢

Morphology

➢

Imaging

➢

Biology

➢

Natural history of each tumor site

➢ But GTV and CTV delineation are independent of the radiotherapy technique used

Planning Target Volume: PTV

• Geometric concept

• Meant to allow for an adequate coverage of the CTV what ever the technique, the movements, the set up uncertainties are

• Volume used for treatment planning

• Volume used for reporting

PTV

Irradiated Volume and Treated Volume: IRV and TV

• IRV : Defined as the volume receiving a significant dose on surrounding normal tissues / Organs At Risk

• Different from the treated volume which is meant to be treated

• Both depend on the technique used

• Both can be evaluated on the dosimetry but IRV evaluation is rather limited by most TPS ➢ Ex: dose estimation outside of the treated field when using non coplanar beams

ICRU 50

ICRU 62 (in addition to ICRU 50)

• Introduces the Conformity Index : CI= treated volume/ PTV

• Recommendations on anatomical and geometrical margins

• Internal Margins : IM are margins integrating physiological movements (breathing, bowel/ rectum/ bladder repletion, swallowing…)

• Internal Target Volume : ITV is defined as the CTV taking into account Internal Margins

Set up Margin: SM

• Margins related to patient positioning:

➢ Positioning uncertainties due to patient external movements ➢ Positioning uncertainties due to body markers ➢ Mechanical uncertainties due to immobilization device precision

• Depend on the technique (ex: tracking) and immobilization material and protocols (ex: thickness of painting markers or tattoos)

What is the definition of the ITV?

A. ITV= GTV + IM B. ITV= CTV + IM C. ITV= PTV + IM D. ITV= GTV + SM E. ITV= CTV + SM F. ITV= PTV + SM

What is the definition of the PTV?

A. PTV= GTV + CTV B. PTV= CTV + IM C. PTV= CTV + SM D. PTV= CTV+ IM + SM

Contouring Guidelines

• Ex: ESTRO breast guidelines

Contouring Guidelines

• Ex: ESTRO breast guidelines

B.Offersen et al radiother oncol 2015

Contouring Guidelines

• Ex: ESTRO breast guidelines

Contouring guidelines

• Anatomical basis are the key!

Contouring guidelines

• Anatomical basis are the key!

ESTRO guidelines

http://www.estro.org/?l=s

Take home messages:

- Inter observer variability in contouring can translate in a systematic error

- Need for a common language: ICRU

- Need for delineation guidelines

- Need for teaching in contouring

What is the weakest point in our modern radiotherapy treatment chain?

A. Dose calculation? B. Positioning uncertainties? C. Contouring uncertainties? D. Quality control of the treatment machine? E. Patient changes (weight loss, movements…)? F. RTTs?

G. Physicists? H. Physicians?

Thank you for you attention

Any question?

How would you score this lecture?

A.Poor B. Sufficient C.Average D.Good E. Excellent

comments can be written on Survey Monkey

Organ at Risk Delineation TARGET VOLUME DELINEATION

Liz Forde (MSc), RTT Assistant Professor Discipline of Radiation Therapy Trinity College Dublin

Sofia Rivera, MD, PhD Radiation Oncology Department

Gustave Roussy Villejuif, France

Advanced skills in modern radiotherapy May 06, 2018

Learning Outcomes

• Discuss the changing roles and responsibilities of RTTs with respect to Organ at Risk (OAR) delineation

• Discuss the impact inaccurate OAR delineation can have on treatment planning

• Discuss the application of dose volume constraints based on delineation protocols

• Identify resources available to support consistency and accuracy in OAR delineation

Why Are OARs So Important?

• Do no harm culture of medicine ➢

Decrease impact of radiation to our patients

• Requirement for inverse planning optimisation process ➢ IMRT ➢ VMAT

• Generates DVH information and assists in prediction of toxicity ➢ Serial and Parallel structures ➢ Assessment of clinical impact and disturbance on daily activities

Why Is Accuracy So Important?

• Consistency and uniformity

➢

Within the department ▪

Prospective data collection

▪ Analysis of local practice and impact on patients

➢

Within the context of clinical trials ▪ Compliance with trial specifications ▪ Allows for collections of data and comparison of outcomes and toxicity at a larger international scale

Why Is Accuracy So Important?

• OAR delineation has significant impact on dose calculation and plan quality in dosimetry

• IMRT and VMAT are inverse planning techniques and as such are driven by volumes ➢ Target and OAR relationship

• Accurate imaging ensures: ➢

Decrease in interobserver variability

➢

DVH calculation

➢

Greater confidence in predicting toxicity

➢ “reduction in inter- and intra-observer variability and therefore unambiguous reporting of possible dose-volume effect relationships” (van der Water, 2009)

Impact on Planning

What is wrong in this picture? What has caused this? What impact would this have?

Possible recommendations put forward by the authors: Contouring by a single user Introduction of MRI into practice Improving the agreement between observers (consensus)

Nelms B et al., Variations in the contouring of organs at risk: test case from a patient with oropharyngeal cancer. IJROBP. 2012; 82(1): 368-378

Question Time!

In my current practice organs at risk are contoured by the:

A. RTT B. Radiation Oncologist C. Medical Physicist D. Dosimetrist

I personally am involved in OAR delineation:

A. Never B. Sometimes C. Frequently D. Always

The New RTT!

“ flexible inter professional boundaries” Schick et al., 2011

“The goal of a radiation therapist undertaking OAR delineation is logical role expansion.” (Schick et al 2011)

The New RTT

• Comparison of practice and confidence • Identified tasks performed at CT Simulation • Results: 84% no change made by RO

Tools for Implementation and Facilitating Change

• Culture of the department ➢ Clinical mentorship ➢

Commitment to evidence based practice

➢

Commitment to role development

➢

Shared goals within the MDT

➢

Open communication

• Prior and ongoing education!

• Even in an ideal environment uncertainties in delineation exist...

Observer Variability in Delineation

• Claude Monet • Photo • 1922

Intra Observer Variability

Inter Observer Variability

Recommendations to Decrease Observer Variability

• Use of contouring guidelines and atlases • Use of secondary imaging data sets • Use of auto-contouring tools ➢

Not to be used in an isolated fashion but to be adjusted for each individual patient

• Attendance at contouring workshops • Multidisciplinary input – open communication • Peer review of contours, regardless of who completed the delineation • Education within the clinic and competency based approach to new roles/responsibilities (Bristow et al., 2014)

Vinod S et al., A review of interventions to reduce inter-observer variability in volume delineation in radiation oncology. JMIRO. 2016, 60(3): 393-406

Auto – Segmentation

• Image content or greyscale method ➢

Appropriate for very high or low contrast structures

• Segmentation without prior knowledge

• Widely available (e.g. flood fill, spindle snake)

• “Common errors include…using the auto-threshold contouring tools in the TPS and not editing the resulting errors” (Gay et al., 2012)

Whitfield G et al., Automated delineation of radiotherapy volumes: are we going in the right direction? BJR. 2013 86(1021): 20110718

Auto – Segmentation

• Atlas based segmentation

• Propagation of segmented structures from an atlas onto the patient image using deformable registration (Lim and Leech, 2017)

• Atlas can be based on: ➢ Single patient dataset ➢

Multiple patient data (based on an average of a range of patients from multiple libraries) ➢ Model based (using library of previously manually contoured patients)

Auto – Segmentation

• Shape model based segmentation

• Concept is extending an active snake approach into an active mesh approach ➢ Driven by greyscale and constrained by shape

Whitfield G et al., Automated delineation of radiotherapy volumes: are we going in the right direction? BJR. 2013 86(1021): 20110718

Auto – Segmentation: Vendor Solutions

Raudaschl P et al., Evaluation of segmentation methods in head and neck CT: Auto- segmentation challenge 2015. Medical Physics. 2017; 44(5): 2020-2036

Auto-segmentation – Beware!

• Attractive due to time saving aspects and support of adaptive RT, but...

• Beware of automaticity!

➢ “Even with the implementation of AS software in the future, it should be reinforced that manual editing is still a necessity for patient safety.” (Lim and Leech, 2017)

➢ “atlas-based automatic segmentation tool ... is timesaving but still necessitates review and corrections by an expert” (Daisne and Blumhofer, 2013)

Question Time!

In your current practice what defines how organs at risk are contoured?

A. In house guidelines

B. Individual preference of the radiation oncologist C. Published consensus guidelines or clinical trials

D. Don’t know

In your current practice how is the small bowel contoured?

A. Individual loops

B. Cavity/space “Bowel bag”

C. Case by case basis

D. It is not contoured

E. Don’t know

Is there Consensus?

eLearning Modules by Experts

QUANTEC

Published Literature

Clinical Trials

Contouring Atlases

So Let’s take a look at the Head and Neck...

Head and Neck

A lot of contouring!

MDT approach!

Critical structures are critical !

Head and Neck

• RTOG Atlases for H&N do not cover OARs!!!

Head and Neck

Available from www.eviq.org.au

eviQ Head and Neck Critical Structures Atlas

• Shows adjacent images with and without contour

• Provides anatomical location, description,

suggested window level and tolerance dose

eviQ Head and Neck Critical Structures Atlas

Note: degradation of image quality due to dental artefact

eviQ Head and Neck Critical Structures Atlas

Remember to view structures in all planes

eviQ Head and Neck Critical Structures Atlas

Remember to use all imaging available for that patient

Published Literature

Consensus panel of Radiation Oncologists from Europe, North America, Asia and Australia

Head and Neck • Don’t worry – even the “experts” have significant inter- observer variability

Head and Neck

• But still worth a read! • Test and table description of anatomy with multimodality images to show

Head and Neck

• Thank you – they have an atlas published as supplementary material

Head and Neck – ESTRO Support

Head and Neck – ESTRO Support

What about clinical trials?

So Let’s take a look at the Pelvis...

AGITG – For Anus

• Bladder ➢

Entire outer wall

• Femoral Heads ➢

Inferior – Cranial edge of the lesser trochanter

• Bowel ➢

Small and large bowel

➢ 15mm superior of PTV down to the rectosigmoid junction • External Genitalia ➢ Male – penis, scrotum, skin and fat anterior to the pubic symphysis ➢ Female - clitoris, labia majora and minora, skin and fat anterior to pubic symphysis • Bone Marrow ➢ Iliac crests, both contoured and combined ➢ Superior - top of the iliac crests ➢ Inferior - superior part of the acetabulum

Take note of positioning at Sim!

RAVES

• Femoral head: ➢

Superior – acetabulum

➢

Inferior – inferior edge of the treatment field

• Bladder: ➢

Whole structure with bulk homogeneity correction for contrast

• Rectum: ➢

Superior – rectosigmoid junction Interior – 15mm inferior to the CTV

➢

PROFIT Trial

• Rectal Wall

• Bladder Wall

• Femoral Head and Neck

Let’s Look at Some Common OARs in the Thorax

Heart

Ribs

Lungs

Spinal Cord

Oesophagus

Brachial Plexus

Main Bronchus

What Are Some of the Challenges You Faced?

• Windowing

• Length to contour

• Contrast

• Motion

• Exclusion of disease

RTOG Thoracic Atlas available from:

http://www.rtog.org/CoreLab/ContouringAtlases/LungAtlas.aspx